In media handling assemblies, particularly in printing systems, accurate and reliable registration of an image as it is transferred is desirable. In particular, accurate registration of an image as it is transferred to a target substrate media or intermediate transfer belt has a direct correlation to image quality.
Contemporary media handling assemblies use controllers, in the forms of automated processing devices, in order to maintain control of the sheets they are handling. Often that control is maintained by adjusting a drive roller or belt velocity which conveys images and/or sheets of paper on transfer belts to a delivery registration datum. A velocity/position command profile is computed by the controller using an algorithm that is designed to deliver the image and/or sheet at a target time to the right place within the system. An actuator commanded by the controller then executes that command profile in order to timely and accurately deliver the image and/or sheet. Such systems are particularly common in printing systems, but are also found in other substrate media handling assemblies.
Typically, control algorithms are employed to analyze the way the system is operating by measuring or monitoring its movement. The control algorithm attempts to control most motion errors, particularly ones that develop slowly and are inherent characteristics of the system itself. One example of such control techniques is feedback control. Feedback control uses a closed-loop system that reacts to inputs and disturbances occurring during system operation. However, as feedback control is reactionary, it tends to lag in its response and thus may not compensate fully for quick or brief disturbances.
Another control technique is the traditional feedforward control, which uses an open-loop system that accumulates information for future use based on the preliminary calibration and/or setup of the system. The feedforward control can eliminate the response lag and anticipate known system disturbances, even quick or brief ones, but it does not respond to exogenous errors, even repeating disturbances that are not part of the initial system calibration.
Those traditional control algorithms do not control errors that occur due to rapidly changing disturbances caused by external influences, such as when a sheet of paper in the normal paper path of the system makes impact with an internal transfer belt, nip assembly or other surface. When the leading edge of a sheet hits a media transfer belt, the nip rollers or similar elements in a system, there is a resultant disturbance that briefly occurs, even though these impacts are anticipated. Similarly, when the trailing edge of that same sheet no longer makes contact with those elements, there can also be a resultant disturbance that briefly occurs. These disturbances are cyclical because as each of a plurality of identical sheets are conveyed through that part of the system, those sheets are not considered part of that system. Particularly since the substrate media can come in different weights, sizes and even material composition. When a novel sheet is introduced to the system, such as, for example, during initialization of a printing machine, when feed trays are changed, and/or when switching between two sheet types, performance of the overall system may change. What is more, the unique characteristics of that novel sheet can change once again with the next print job, that uses an even different substrate media.
The unique disturbances caused by a particular set of sheets on the system are considered exogenous since those sheets are not considered part of the media handling assembly. Such exogenous disturbances are not fully compensated for by feedback control or contemporary feedforward control systems. Also, such exogenous disturbances are not consistent between different substrate media, making them somewhat unique for each set.
Accordingly, it would be desirable to provide a system and method capable of more accurately reducing registration errors in a media handling assembly, and thereby overcomes the shortcoming of the prior art.